COX-2抑制剂的定量构效关系研究

Quantitative Structure-activity Relationship Study of COX-2 Inhibitors

选择性抑制环氧酶-2可以达到消炎镇痛的目的。本文应用支持向量回归机、k最近邻以及C4.5决策树三种机器学习方法建立了COX-2抑制剂的回归模型,分析其定量构效关系。本文又对三个模型的预测效果进行了评价,误差均能满足精度要求。最后,本文比较了它们的优缺点和适用范围。

Selective inhibition of cyclooxygenase-2（COX-2） can reduce inflammation and pain. In this work, three machine learning methods（support vector regression, k-nearest neighbor and C4.5 decision tree） were applied to establish the regression models of COX-2 inhibitors and analyze the quantitative structure-activity relationship. The predictive effects of the three models were also evaluated, all showing low forecast errors. Finally, their advantages, disadvantages and the range of their applications were compared.